Recently over on the Fediverse I mentioned I'd broken Ctrl-D toadd new bookmarks in my Firefox :

TIL that I accidentally broke bookmarking on my primary Firefox. Thisis actually convenient because the only time I bookmark somethingon it is when I absentmindedly used Ctrl-D to try to close thewindow/tab. (My terminals close with Ctrl-D, my Firefox windows withCtrl-W, I have a bunch of both, yes life is fun.)

One of the irritating ongoing changes in

If you hover the mouse over a link in Firefox (or pretty much anyother browser), the browser will show you the URL of the link targetso you can see where you'll wind up if you click the link; this URLtarget appears down at the bottom of the page. Well, in theory,since there are a lot of things that can happen in between yourclick on a link and where you wind up. In Firefox, as far as I knowthat URL target is rendered in

In a comment on my entry on how you should automate some basic restoretesting of your backups , Simon made aperfectly reasonable suggestion:

Another relatively basic, but useful, check is to do some plausibilitycheck on the backup size. For most systems huge jumps in backup size(in either direction) likely mean you are not backing up what youare thinking. This is a bit more complicated than what you mentionin your article, since it needs some tuning and can generate falsepositives. But I think it still

Recently on the Fediverse, Ben Zanin said this important thing :

Yesterday was world backup day!

Please remember that the value of your backups is determined by thecadence of your successful restoration tests.

Please also remember that when backing up your enterprise applicationsyou need to actually follow vendor admonitions about synchronizingyour database and filesystem snapshots, keeping your backups across anair gap, and never connecting your system network to your managementnetwork. These are not someone else's problems, they are yours.

💖

Everyone says that you don

I imagine that you've heard of HTTP/3 by now; as Wikipedia describesit, it's "the third major version of the Hypertext Transfer Protocolused to exchange information on the World Wide Web". I'm generallya quiet enthusiast of adopting new HTTP things when we can (we'vebeen reasonably prompt to add HTTP/2 to our Apache web serverswhen it was possible), but with HTTP/3 I think we're likely to takea much more cautious and slow

I've recently been looking into how people add Multi-FactorAuthentication (MFA) to their OpenVPN systems, both using commercialsolutions and home grown ones. One of the things that makes thisdifficult is that I believe the OpenVPN authentication protocol isold fashioned enough that it doesn't provide for multi-stepinteraction. Instead, clients can send either or both of a TLS clientcertificate and a username plus password pair to the server, andthe server gets to decide. However, common OpenVPN server softwareallows

For reasons outside the scope of this entry, I've recently beengiving Gnome-Terminal more of a try as my secondary terminal program,instead of urxvt (my primary terminal program remains xterm). Gnome-Terminal is a perfectly fine default terminal program and so most of this has gone perfectlywell. But the more I used gnome-terminal, the more I found myselfhaving problems with how by default it runs the text almost rightup against the edge of the window (on the left) and the

Over on the Fediverse, I shared a discovery from our mail logs :

In the 'that's not how you do it' category, spotted in our email rejectlogs today:

Content-Transfer-Encoding: amazonses.com

(We rejected it for having an absurdly long line, over 200,000 bytes,which appears to have been almost all of the message.)

The MIME Content-Transfer-Encoding header is supposed to tell you the encoding of the MIME part inquestion,

On the Fediverse, Alan Coopersmith recently shared this :

@bsmaalders @cks writing a temp file and renamingit also avoids the failure-to-truncate issues found in screenshotcropping tools recently (#aCropalypse), but as some folks at workrecently discovered, you need to be sure to fsync() before the rename,or a failure at the wrong moment can leave you with a zero-lengthfile instead of the old one as the directory metadata can get writtenbefore the file contents data on ZFS.

On